In Augmented Apps, we examine how product teams are exploring AI and Machine Learning to make their products more intuitive and enhance user experience. 

As a product owner or manager, you are constantly faced with pressure to be innovative and deliver value and great experiences to your users. Innovation means different things in different industries, but one common feature that many apps are gravitating toward is analytics. From ride-sharing and food delivery to running apps, language learning, and even subscription services, users want data. They want to know how they’re using your service, what it’s doing for them, and how they could be doing it better (running faster, speaking more Spanish, etc.).

Delivering embedded insights is an important step in providing additional value to your customers today. If your application doesn’t already deliver embedded insights, then you’re not just missing out on an opportunity to delight customers and create additional revenue and growth for your company, you’re putting your company’s continued success at risk. In this post, we’ll explore why embedded analytics are so vital and new ways to let users actually talk to their data.

Building beyond basic embedding

Assuming that you have embedded analytics, let’s think about how you can take them one step further. Simply delivering predetermined visualizations and dashboards embedded within your application, while relevant and useful, has become basic table stakes. It’s not new anymore! Also, it might take some time to navigate, especially for a non-experienced user. You need to move beyond the typical and provide innovative experiences, preferably without spending a ton of time and effort building them from scratch.  

Additionally, under the current embedded analytics paradigm, it’s incumbent upon you and your team to think up every possible question a customer might want to know about and pre-build the answers into your product’s embedded analytics.

The problem with this process is immediately apparent: You and your teammates can’t possibly cover all scenarios and variations of questions that your consumers could come up with. Now you’re in the unenviable position of having to cater to a larger audience by delivering deeper insights when even the users themselves don’t necessarily know all the questions they want to ask yet. How do you avoid being a bottleneck to your customers? How do you democratize insights?

Natural language querying: Your key to democratizing deeper insights

Wouldn’t it be great if you could enable your customers to simply ask a question in natural language and have your application spit out an insight in response? That’s called natural language querying (NLQ), and it’s a natural outgrowth of natural language processing (NLP) and generation (NLG), two machine learning applications that allow computers to understand and communicate in human speech, spoken or written. NLQ is a powerful tool in giving your users greater control over how they interact with the data in your app without you and your team having to come up with every possible question and drill-down on your own.  

Eager to deploy NLQ in your own product? Now it’s easier than ever.

Sisense natural language queries can be embedded directly into any application to easily bring the power of NLQ to your customers without you having to leverage data scientists at your company or go out and master machine learning skills. Delight your customers and end-users with cutting-edge AI-driven insights delivered right in the context of their workflows. 

Ask questions, get insights

Sisense NLQ enables users to ask questions in natural language and get insights and visualizations back, based on specialized machine learning/NLQ algorithms. The NLQ data model is created automatically based on your app’s dashboard content and is versatile enough to be modified manually by adding or removing elements to suit your audience’s needs. If an end-user is not savvy with the data (model), then Sisense NLQ has them covered with synonyms, spelling corrections, word auto-complete and type-ahead recommendations support out of the box. Sisense NLQ also responds to questions seeking forecasts and trends on the data (by integrating with Sisense Forecasting and Trends). 

Want an example of how it works? Imagine this scenario: Your customer or end-user is using your application (an IT incident management application), and they want to understand the number of critical incidents opened in the last hour. Instead of searching for this stat in a corresponding dashboard (if it even exists) or asking the design team to build it, they can simply type in the question within their application and have Sisense NLQ run behind the scenes to give them an answer. This saves everyone time and aggravation and ultimately leads to faster data-driven decisions. 

Choose your own embedded NLQ design strategy

No matter what your embedded analytics design is, embedded NLQ can be easily integrated via any of Sisense’s embedding capabilities (iFrame embedding, Embed SDK, or Sisense.JS) to meet your specific use case and needs.  

One way you can easily present users with embedded NLQ functionality is by setting a trigger for the “Simply Ask” pop-up in the host application:

Simply Ask

Or you can embed the Simply Ask capability as a standalone feature in your application. Alternatively, you can enable Simply Ask as a part of the embedded Sisense application.

Deliver unique experiences

As a product leader, you’re called upon to be an innovator. Democratizing insights for your customers in context and within your application is an important step in bringing data closer to your customer. With Sisense NLQ and the power of Sisense embedding, you can improve your product adoption, retention, and satisfaction through innovative insight delivery, without requiring new dev resources or help from other teams.

This is just the beginning: Apps and devices are starting to understand natural language instructions better. Take a leap with your current offering using embedded NLQ, then look for more ways to teach your apps to talk, give users the data experiences they crave, and more. The time to be bold is now; what that looks like is up to you.

Adding AI to Products

Shruthi Panicker is a Sr. Technical Product Marketing Manager with Sisense. She focuses on how Sisense can be leveraged to build successful embedded analytics solutions covering Sisense’s embedding and customization capabilities, developer experience initiative, and cloud-native architecture. She holds a BS in Computer Science as well as an MBA and has over a decade of experience in the technology world.

Tags: | |